从神经网络决策到训练案例：基于案例决策理论的精确阐述
Do Neural Networks Preserve Case Structure? Case-Based Decomposition, Interpretation, and Decision Consistency
浏览论文内容
中文总结 AI 辅助
研究神经网络决策，基于案例决策理论，通过OLS行动读出的精确案例分解，能追溯分数到训练案例、衡量行动一致性等，在多任务中恢复案例级偏好结构，实现高一致性且审计简单，探测器保真度由分数重建衡量。
中文摘要 AI 辅助
神经网络在医疗诊断、信贷审批和能源投标等高风险领域越来越多地指导决策。这些场景中的审计需要案例级证据，即哪些训练案例支持某个行动以及它们带来了什么结果。基于案例的决策理论（CBDT）通过汇总记忆案例的结果支持来形式化这种推理。我们表明，在固定神经表示上拟合的OLS行动读出允许精确的基于案例的分解。每个行动分数是训练案例回报的加权和，系数由经验Gram几何确定。我们确定了CBDT相似性语义的充分条件；在其之外，系数通常应被视为带符号的Gram几何影响。这种分解产生了审计信号，可将分数追溯到训练案例、衡量行动一致性并识别薄弱支持。在合成CBDT、PJM、成人收入和违约信贷任务中，该方法恢复了案例级偏好结构，并在比较的归因基线中实现了最高的平均前30一致性，同时在支持重建方面保持竞争力。审计只需要拟合一个OLS顶层探测器，而无需重新训练表示或访问原始优化轨迹；探测器保真度通过分数重建来衡量。
英文摘要
Neural networks increasingly inform consequential decisions, making their reliability increasingly important. Yet their internal mechanisms provide little evidence of whether decisions remain grounded in the training cases and which cases ultimately support or oppose their outcomes. Without this connection between decisions and training cases, users cannot determine whether a model has learned reliable decision patterns from data. This motivates a fundamental question: do neural networks preserve case structure? We establish a connection between neural networks and Case-Based Decision Theory (CBDT), showing that trained neural networks can preserve a recoverable case structure through their learned representations. Such a structure allows fitted decision margins to be decomposed into individual case contributions. We identify the conditions under which this recovered case structure admits a CBDT interpretation. We further establish decision consistency between this interpretation and the decision selected by the original neural network. Experiments on a controlled CBDT setting and three decision tasks based on real-world data validate our approach. These results connect neural network decisions with the cases that shape them. This connection allows model choices to be traced back to supporting and opposing cases, providing a basis for assessing the reliability of neural network decisions.
发表机构
- Huazhong University of Science and Technology（华中科技大学）
机构由 AI 辅助整理，请以论文原文为准。
